Comparison of neural network models for fault diagnosis of power systems

Abstract This paper compares the performance of two neural network (ANN) models for fault diagnosis of power systems. Radial basis function and back-propagation networks are compared with reference to generalization, training time and number of training patterns needed for each model.